def setup(self, stage: str = None) -> None:
print(f"IAMSyntheticParagraphs.setup({stage}): Loading trainval IAM paragraph regions and lines...")
if stage == "fit" or stage is None:
line_crops, line_labels = load_line_crops_and_labels("trainval", PROCESSED_DATA_DIRNAME)
X, para_labels = generate_synthetic_paragraphs(line_crops=line_crops, line_labels=line_labels)
Y = convert_strings_to_labels(strings=para_labels, mapping=self.inverse_mapping, length=self.output_dims[0])
transform = get_transform(image_shape=self.dims[1:], augment=self.augment) # type: ignore
self.data_train = BaseDataset(X, Y, transform=transform)
def __init__(self):
data = IAMParagraphs()
self.mapping = data.mapping
inv_mapping = data.inverse_mapping
self.ignore_tokens = [
inv_mapping["<S>"], inv_mapping["<B>"], inv_mapping["<E>"],
inv_mapping["<P>"]
]
self.transform = get_transform(image_shape=data.dims[1:],
augment=False)
with open(CONFIG_AND_WEIGHTS_DIRNAME / "config.json", "r") as file:
config = json.load(file)
args = argparse.Namespace(**config)
model = ResnetTransformer(data_config=data.config(), args=args)
self.lit_model = TransformerLitModel.load_from_checkpoint(
checkpoint_path=CONFIG_AND_WEIGHTS_DIRNAME / "model.pt",
args=args,
model=model)
self.lit_model.eval()